[cpp-threads] RE: "Agenda" for august 23-25 concurrency meeting

[Boehm, Hans]

Just to clarify, here is my current draft agenda, with corrected dates,
and some requested additions:

Starting at 9:00, Wednesday, Aug. 23:
Atomics
	Two levels of atomics?
	Bottom level operates on plain scalars, volatiles, or special
data?
		Does backward compatibility matter?
	double-word CAS?
	Does default atomic assignment include trailing fence?
		I.e. does it behave like Java volatiles?
	Other details?

Wednesday after lunch talk:
	OpenMP status overview (Volunteer TBD)

Starting Wednesday, around lunch?
Memory model
	Fairly detailed presentation (Boehm and/or Nelson)
	Construction of statics thread-safe by default?
		A tentative decision would be nice.
	Treatment of infinite loops.
	Do ordered atomics effectively order "raw" (unordered)
		atomic operations?
		(See
http://www.decadentplace.org.uk/pipermail/cpp-threads/2006-August/000989
.html) 
	Other issues?

Thursday 8/24 after lunch talk:
	Intel Threads Building Blocks (Arch Robison)

Starting Thursday afternoon?
Threads API
	Should we be focusing on threads?
		As opposed to processes withshared memory.
	Approach?
		I (Hans) think the top level alternatives are
			A) Boost + modifications
			B) More drastic redesign
				With or without added syntax.
			C) fallback: something minimal to let us talk
			about threads in the standard while using
			Boost/pthreads/win32/... in real programs
	Issues that are probably open either way:
		Is there a C-compatible standard API?
		Constructor/destructor handling.
		Exception propagation past joins?
		Any cancellation support?
	Probably many other issues.

Probably starting Friday 8/25 after lunch:
Library Issues
	Can we start a list?
	Can we leave basic_string as is?
	Do I/O operations introduce synchronizes-with relationships?
	Which I/O operations are atomic?
	What about thread-specific library extensions such as Intel TBB?
	Who is willing to look at these issues?
	Probably many other issues.

Close at around 4:00 p.m.

Recommended prereading (Hans' biased priority ordering):
	Draft of the Nelson & Boehm memory model/sequencing paper which
we hope
	to get out by Friday or so.  If you can't wait until then, the
current
	strawman memory model proposal.
	Herb's Prism paper.
	N2047 atomics proposal.
	Pete Becker's thread library proposal (N1907).
	Except for Herb's paper, these are/will be all referenced from
	http://www.hpl.hp.com/personal/Hans_Boehm/c++mm/.
	A summary of the other threads API discussion at the Berlin
	C++ meeting (Ugly version at
	
http://www.decadentplace.org.uk/pipermail/cpp-threads/2006-April/000928.
html)
	Anything else that's referenced from that page and you find
	interesting.